received 20.05.2014 accepted 06.11.2014 Bibliography DOI http://dx.doi.org/ 10.1055/s-0034-1395637 Published online: December 12, 2014 Horm Metab Res 2015; 47: 225–231
© Georg Thieme Verlag KG Stuttgart · New York ISSN 0018-5043 Correspondence A. Salvadori
Istituto Auxologico Italiano P. O. Box 90 28921 Verbania (VB) Italy Tel.: + 39/0323/514 212 Fax: + 39/0323/514 411 [email protected] Key words ●▶ leptin ●▶ insulin resistance ●▶ obesity ●▶ physical training ●▶ anaerobic threshold
Leptin Level Lowers in Proportion to the Amount of
Aerobic Work After Four Weeks of Training in Obesity
cise in relation to energy expenditure [11]. In normal as well as in obese subjects, a period of prolonged physical training also promotes a decrease in leptin [12, 13].
Regular exercise is an important strategy in the management of obesity, together with hypoca-loric diet and appropriate lifestyle. The aim of this study was to elucidate in obesity the behav-iour of leptin after a comparable type of training at work loads of different intensity, that is: 1) after exclusively aerobic work, and 2) after aerobic work plus a bout of anaerobic work. On the basis of previous observations on GH and nonesterified fatty acids (NEFAs) [14, 15], we might presume different behaviours of leptin with regard to differences in quantity of physical training.
Subjects and Experimental Protocol
▼
Among 50 initially screened patients admitted between January 2009 and December 2011 to Auxologic Italian Institute for workup of obesity and weight management program, 16 subjects (8
Introduction
▼
Leptin is a hormone mainly secreted by adipocytes that works to regulate body weight and satiety [1]. Its levels in blood serum are higher in obese com-pared to lean subjects, but the significance of this difference on the development of obesity remains uncertain. In fact, the failure of the metabolic response to endogenous leptin in the setting of obesity, a condition also known as leptin resistance, is a question still open to debate [2].
Previous studies in different experimental condi-tions have pointed out that serum leptin may decrease in state of acidosis [3], elevated glucose uptake in the presence of lactate [4], and inhibi-tion of glycolysis [5]. In contrast, a stressed leptin secretion by adipocytes has been shown after acute administration of growth hormone (GH) [6], glucose infusion [7], and acute infusion of glucosamine with an increase in leptin mRNA [8, 9].
Acute physical exercise promotes a decrease in plasma leptin levels in trained subjects, as observed 9 h after resistance exercise [10], or at the end of a moderately intense prolonged
exer-Authors A. Salvadori1, P. Fanari1, A. Brunani2, P. Marzullo3, 4, F. Codecasa1, I. Tovaglieri1, M. Cornacchia1,
P. Palmulli1, E. Longhini1
Affiliations Affiliation addresses are listed at the end of the article
Abstract
▼
Leptin values are higher in obesity. Physical exer-cise reduces fat mass (FM) and decreases leptin levels. Intensity of physical training seems to play a role in reducing circulating leptin. In 16 obese subjects (8 men and 8 women, age 38.6 ± 3.9 years, BMI 35.9 ± 1.8 kg/m2), leptin was
sampled before and after 4 weeks of controlled training. Eight subjects (4 men and 4 women) performed an aerobic training schedule (Group A), the remainders an aerobic training program with a bout of work beyond the anaerobic thresh-old (AT) (Group B). Training determined a reduc-tion in leptin levels in both groups, which was significant in Group A (12.2 vs. 27.8 μg/l, p < 0.05),
even when related to the change in FM (0.372 vs. 0.762 μg/l/kg, p < 0.05). FM decreased signifi-cantly in Group B when compared to Group A (–7.4 vs. –2.6 kg, respectively, p < 0.001). While in Group A the slight loss of FM was aggregated to a significant decrease in leptin levels, the opposite occurred in Group B. In Group A, leptin lowering was proportional to the amount of total work performed (p < 0.001, R2 = 0.89). In obesity, a
reduction is observed in leptin levels after short-term training, which is seemingly dissociated from concomitant decrease of FM. Aerobic train-ing alone appears to be linked to a greater leptin reduction, which is well correlated with the amount of work performed.
males and 8 females, age 38.6 ± 3.9, range 22–59 years, body mass index (BMI) 35.9 ± 1.8 kg/m2, range 31–43) were enrolled
and studied at the beginning and at the end of a 4-week inpa-tient period.
Exclusion criteria were smoking, physical inability, arterial hyper-tension, diabetes, or any other cardiovascular or metabolic disor-der as well as thyroid and autoimmune disordisor-ders. Women were analysed during early follicular phase of the menstrual cycle. Anthropometric measures after voiding, EKG, routine blood and urine analysis were carried out to exclude medical illnesses. Bio-impedance analysis (BIA 101/S Akern, Florence, Italy) previously validated [16] to exclude fluid overload and fat mass (FM, kg) and fat-free mass (FFM, kg) was performed in each participant before and after the training period. A preliminary nocturnal O2
saturation monitoring excluded the presence of significant periods of pathological low values in each participant.
The protocol was approved by the Ethics Review Committee of the Institute and written informed consent was given by all the subjects before participation, in accordance with the Helsinki Declaration of the World Health Organization (1964, amended in 1975 and 1983) and with the Updated Ethical Standards in Sport and Exercise Science Research [17].
Body weight was stable in each subject for at least 3 months before enrollment. Prior to exercise testing, subjects were asked to restrain from strenuous activity for at least 24 h. Following an overnight fast, in the morning every subject performed a con-tinuous incremental test on a Gould bicycle ergometer with power output increased by 20 Watt (at 60 rpm) every 4 min until the pedalling frequency could no longer be maintained despite verbal encouragement. Subsequently, everyone continued ped-alling effortlessly for 2 min, and then stopped and remained in a sitting position for up to 30 min.
The Vmax 229 (Sensor Medics, Yorba Linda, CA, USA) gave con-tinuous analysis of oxygen consumption (VO2), CO2 production
(VCO2), and ventilation (VE). Work capacity was assessed by the
maximum work rate obtained together with the detected VO2
max related to predicted VO2 max [18], and ventilatory
anaero-bic threshold (AT) measured by the V-slope method [19]. Heart rate (HR) and EKG signals were recorded by a CASE 6.5 (GE Med-ical System Milwaukee, WI, USA), and oxyhaemoglobin satura-tion was controlled by a Radiometer percutaneous oxymeter every 20 s. Calibrations were performed prior to each test. Sampling was carried out before the test from the antecubital vein of the right or left arm and analysed for determination of leptin (Linco’s RIA; sensitivity, 0.5 μg/l; intra-assay CV, 8.3 %), nonesterified fatty acids (Randox Laboratories USA; sensitivity, 0.072 mmol/l; inter- and intra-assay CV, 4.5 and 4.74 %, respec-tively), insulin (Immulite 2000 Analyzer; Diagnostic Products Corp., Los Angeles, CA; sensitivity, 2 μlU/mL; inter-and intra-assay CV, 4.0 and 5.1 %, respectively), and glycaemia (Gem Pre-mier 3000; Instrumentation Laboratory). In addition to absolute concentrations leptin was expressed as values normalised for fat mass (leptinFM) both at the beginning and at the end of the study,
and variations between these values (leptinFM after training/
leptinFM before training) as percent. The modified HOMA model
was used to yield an estimate of insulin sensitivity and β-cell function from fasting plasma insulin and glucose concentration [20].
After the preliminary exercise testing, 8 subjects were consid-ered for Group B; the others were considconsid-ered for Group A and an effort was made to maintain an equal distribution between the 2 Groups as far as sex, age and BMI are concerned.
Group A performed an exclusively aerobic training (power out-put attaining 70 % of HR registered at AT: 30 min, 2 sessions daily, 6 days/week, 4 weeks). Group B performed a mixed training (aerobic part attaining 70 % of HR registered at AT: 25 min, fol-lowed by anaerobic part attaining 85 % of HR max: 5 min, 2 ses-sions daily, 6 days/week, 4 weeks). About Group B we reasoned that, in accordance with our previous experience, 5 min of work beyond anaerobic threshold could be performed by a great majority of obese subjects. Planning 5 min 2 times a day for 6 days/week for 4 weeks, a total amount of 4 h of anaerobic work would have been performed. This might represent a significant quantitative difference from aerobic work alone.
During the training period increments of power outputs were imposed in both groups in view of the actual reconditioning and the aforesaid HR were maintained as the reference standards. At the end of each training period, everyone underwent a conclu-sive ergospirometric test according to the same protocol setting as the entry test and sampling.
During the study every subjects observed a balanced diet (57 % carbohydrate, 25 % lipid, 18 % protein) corresponding to their basal metabolic rate determined by indirect calorimetry with a ventilated canopy (Vmax 229, Sensor Medics, Yorba Linda, CA, USA), and no strict dietary restriction was imposed.
Statistical analysis
ANOVA was employed to determine differences between groups and between pre- and post-training values with a statistical power at least higher than 70 % [21]. We used the least-squares criterion applying ANOVA to the regression model to calculate the straight-line multiple regression [22]. Values were expressed as mean ± SEM. A difference was considered statistically signifi-cant for p < 0.05.
Results
▼
Group A consisted of 8 subjects (4 M, 4 F) with a mean age of 39.6 ± 3.7 years, Group B consisted of 8 subjects (4 M, 4F) with a mean age of 37.8 ± 4.1 years. Anthropometric data are summa-rised in ●▶ Table 1. Basal BMI, weight and FM were not
signifi-cantly different between the 2 groups (p = NS, p = 0.051, and p = NS, respectively), while height and FFM were significantly lower in Group A (p = 0.039 and p = 0.044, respectively) by 2 tailed analysis of variance ( ●▶ Table 1). Training did not
signifi-cantly decrease body weight compared to baseline both in Group A (92 ± 8 kg vs. 89 ± 7 kg) and Group B (114 ± 8 kg vs. 107 ± 7 kg). Nevertheless, the decrease in FM resulted significantly higher in Group B compared with Group A (–7.4 ± 1.1 kg vs. –2.6 ± 0.8 kg,
Table 1 Anthropometric and functional data of the samples.
Variable† Group A Group B
Before training After training Before training After training Weight (kg) 92 ± 8 89 ± 7 114 ± 8 107 ± 7 Height (cm) 162 ± 4 – 176 ± 4 – BMI (kg/m2) 35 ± 2 34 ± 2 37 ± 2 35 ± 2 Fat-free mass (FFM) (kg) 53 ± 4 53 ± 4 68 ± 6 68 ± 6 Fat mass (FM) (kg) 39 ± 4 36 ± 3 46 ± 3 38 ± 3 Exercise peak (Watts) 95 ± 9 109 ± 10 125 ± 9 138 ± 10 Anaerobic threshold (AT) (Watts) 67 ± 6 76 ± 6 86 ± 7 94 ± 9
† Values are mean ± SEM. BMI: Body mass index
p < 0.001), even when corrected for difference in starting FM (–16.6 % ± 3.0 vs. –7.1 % ± 1.3, p < 0.05, respectively).
●▶ Table 2 shows the individual modifications of weight, FM, and
FFM of both groups after training.
The total amount of physical work performed during the study period resulted significantly higher in Group B than in Group A (mean 6 240 ± 510 Kjoules vs. 4 270 ± 330 Kjoules, p = 0.018, respectively) ( ●▶ Table 3). After training periods, work capacity
improved albeit not significantly both in Group A and Group B ( ●▶ Table 1).
Heart rate increased in both groups during exercise testing in accordance with the increase of work output either before and after the training period ( ●▶ Table 3), nevertheless after training
females of Group A seem to evidence a more pronounced increase in correspondence of peak activity.
In ●▶ Table 4, 5 are reported single session work outputs at the
beginning of training and at the end of training in Group A and Group B as a whole and as a ratio to FFM, and, within brackets, the share of anaerobic work of each participant of Group B. In ●▶ Table 5, with regard to the differences in FFM and
perfor-mance between the 2 groups, we have added the calculated findings of simulated equal work outputs obtained by replacing the performed anaerobic shares of each subject with the hypo-thetical corresponding shares of aerobic work. It is possible to observe that the singular initial session performed by Group A and the initial simulated session of Group B do not differ when
expressed as ratio to FFM (1.53 vs. 1.58 Kjoules/FFM, respec-tively).
Similarly, the total work performed by Group A does not signifi-cantly differ from both simulated and real total work performed by Group B (81.6 ± 4.2 vs. 82.6 ± 5.3 Kjoules/FFM and 81.6 ± 4.2 vs. 93.0 ± 5.5 Kjoules/FFM, respectively).
A decline in leptin levels, both when expressed as absolute val-ues and in relation to FM, reached statistical significance only in Group A, while it decreased nonsignificantly in Group B ( ●▶ Table
6). Individual values of leptin before and after training and the
variations between these values (leptinFM after training/leptinFM
before training) as percent are reported in ●▶ Table 2. In Group A,
these findings were paralleled by a nonsignificant reduction of insulin levels after training, while these were unchanged in Group B ( ●▶ Table 6). Fasting plasma glucose was unmodified in
either groups at the study end. After training HOMA2-B was sig-nificantly lower in Group A ( ●▶ Table 6). NEFA significantly
decreased in Group A and significantly increased in Group B after training ( ●▶ Table 6).
In Group A, we found a linear inverse correlation between the total work performed and the ratio of leptinFM after training and
leptinFM before training in percent (R2 = 0.89) ( ●▶ Fig. 1) which
remains significant even after the addition of NEFA and insulin variations (adjusted R2 from 0.847 to 0.774). In Group B, after
excluding one outlier (marked with * in ●▶ Fig. 1), a linear direct
correlation was documented (R2 = 0.88). Positive values were
Table 2 Individual and mean variations in weight, FM, FFM, and leptin together with (leptinFM after/leptinFM before) percent in men and women of Group A and Group B.
Group A (LeptinFM after/
leptinFM before) %
Males Weight (kg) Fat mass (kg) Fat-free mass (kg) Leptin (μg/l)
Before After Before After Before After Before After
1 117.4 113.0 38.3 34.2 79.1 78.8 63.0 6.2 –89.0 2 131.0 130.0 59.0 57.1 71.8 72.9 27.2 12.1 –54.0 3 79.7 76.1 35.7 33.7 44.0 42.4 25.4 16.7 –30.0 4 78.9 77.0 30.9 30.4 48.0 46.6 10.8 7.6 –28.6 mean ± SE 101.7 ± 13.2 99.0 ± 13.4 41.0 ± 6.2 38.8 ± 6.1 60.7 ± 8.7 60.0 ± 9.2 31.6 ± 11.1 10.6 ± 2.4 –50.4 –2.7 % –5.4 % –1 % –66.5 % Females 5 76.5 73.6 30.6 26.4 45.9 47.2 10.3 6.4 –28.2 6 69.9 65.5 27.1 24.7 42.8 40.8 16.4 10.2 –31.8 7 82.2 80.0 37.1 35.2 45.2 44.8 35.4 14.8 –56.0 8 98.0 96.8 52.9 48.5 45.1 48.3 33.9 23.8 –23.4 mean ± SE 81.6 ± 6.0 79.0 ± 6.6 36.9 ± 5.7 33.7 ± 5.4 44.8 ± 0.7 45.3 ± 1.7 24.0 ± 6.3 13.8 ± 3.7 –34.9 –3.2 % –8.7 % + 1.3 % –42.5 % Group B
Males Before After Before After Before After Before After
1 141.0 137.3 56.5 47.1 84.5 89.6 22.0 26.2 + 44.0 2 * 147.3 139.1 50.3 44.8 97.0 93.7 9.1 3.7 (–54.0) 3 123.0 115.3 50.0 41.0 71.0 73.7 12.6 12.6 + 21.8 4 98.6 94.3 36.2 31.3 62.4 62.4 48.5 26.8 –38.9 mean ± SE 127.5 ± 10.9 121.5 ± 10.6 48.2 ± 4.9 41.0 ± 3.5 78.7 ± 7.6 79.8 ± 7.2 23.1 ± 8.9 17.3 ± 5.6 (–6.8) –5.1 % –15.4 % 0 % –21.1 % Females 5 100.7 96.1 51.8 44.2 48.9 51.3 29.1 27.0 + 8.7 6 84.0 70.0 33.2 20.6 50.8 48.8 34.0 25.3 + 19.9 7 106.8 100.6 53.1 47.0 53.7 53.0 32.4 22.9 –20.2 8 110.3 101.7 38.1 31.4 72.2 69.7 5.6 3.6 –21.8 mean ± SE 100.4 ± 5.8 92.1 ± 7.5 44.0 ± 5.0 35.8 ± 6.1 56.4 ± 5.4 55.7 ± 4.7 25.3 ± 6.6 19.7 ± 5.4 –3.4 –8.3 % –18.6 % –1.3 % –22.2 %
* The outlier value of leptinFM after/leptinFM in percent for the male subject 2 is given in parentheses
detected in correspondence to maximal training work outputs in 3 of 4 cases. When both groups were merged, such correla-tions were lost.
We have looked for any multiple correlation between leptin var-iations and the tested variables in both groups (Kjoules per-formed, FFM, NEFA, insulin, and glucose), finding no significant correlation among them.
Discussion
▼
In this study, we looked at leptin dynamics in 2 groups of obese subjects after a short period of physical training at different work out-puts. The interest in leptin secretion is due to the effect of this hormone on satiety control in which it plays an important role, together with other adipokines, as well as
pan-Table 3 Individual heart rate and work outputs (Kjoules) during exercise testing before and after the aerobic training period in men and women of Group A and the aerobic plus anaerobic training period of Group B.
Group A
Heart rate before training Heart rate after training Kjoules
Males Rest AT Max 70 % AT Rest AT Max 70 % AT
1 78 147 168 103 74 121 146 85 6 048 2 80 138 168 97 87 142 176 99 5 140 3 80 121 143 84 72 126 156 88 3 326 4 85 132 154 92 81 129 150 90 3 629 mean ± SE 81 ± 1 134 ± 5 158 ± 6 94 ± 4 78 ± 3 129 ± 4 157 ± 7 90 ± 3 4 536 ± 641 % vs. rest + 66 + 96 + 16 + 65 + 100 + 15 Females 4 80 128 148 89 86 138 168 96 3 931 5 81 124 134 87 68 121 150 84 4 234 6 65 129 147 90 75 126 146 88 4 536 7 97 150 164 105 96 152 166 106 3 326 mean ± SE 81 ± 6 133 ± 6 148 ± 6 93 ± 4 81 ± 6 134 ± 7 157 ± 6 93 ± 5 4 007 ± 258 % vs. rest + 66 + 86 + 16 + 66 + 96 + 16 Group B
Heart rate before training Heart rate after training Kjoules
Males Rest AT Max 70 % AT 85 % Max Rest AT max 70 % AT 85 % Max
1 71 120 152 84 129 71 116 148 81 126 7 128 2 67 114 148 80 126 64 108 160 76 136 8 878 3 82 121 152 85 129 81 127 158 89 134 7 380 4 64 110 142 77 121 70 112 147 78 125 4 997 mean ± SE 71 ± 4 116 ± 4 148 ± 2 81 ± 2 126 ± 2 71 ± 4 116 ± 4 153 ± 3 81 ± 3 130 ± 3 7 096 ± 799 % vs. rest + 64 + 110 + 16 + 71 + 62 + 116 + 13 + 83 Females 5 76 122 155 85 132 72 125 156 87 133 6 257 6 77 120 150 84 127 76 122 152 85 129 4 874 7 88 136 168 95 143 86 130 158 91 134 5 249 8 82 135 163 94 139 84 133 171 93 145 5 126 mean ± SE 81 ± 3 128 ± 4 159 ± 4 89 ± 3 135 ± 4 79 ± 3 127 ± 2 159 ± 4 89 ± 2 135 ± 3 5 376 ± 304 % vs. rest + 59 + 97 + 11 + 67 + 61 + 101 + 13 + 70
AT: Anaerobic threshold
Table 4 Individual work performed at beginning and ending training of Group A. Group A
Single session work outputs at the beginning of training Single session work outputs at the end of training
Males Watts Kjoules Kjoules/FFM Watts Kjoules Kjoules/FFM
1 2 100 126.0 1.59 2 100 126.0 1.58 2 1 470 88.2 1.26 2 100 126.0 1.75 3 1 050 63.0 1.43 1 260 75.6 1.78 4 1 260 75.6 1.58 1 260 75.6 1.62 mean ± SE 1 470 ± 227 88.2 ± 13.6 1.47 ± 0.08 1 680 ± 242 100.8 ± 14.5 1.68 ± 0.05 Females 5 1 260 75.6 1.61 1 470 88.0 1.86 6 1 260 75.6 1.77 1 680 101.0 2.42 7 1 260 75.6 1.64 1 890 113.0 2.47 8 1 050 63.0 1.35 1 260 75.6 1.53 mean ± SE 1 208 ± 53 72.5 ± 3.2 1.59 ± 0.09 1 575 ± 136 94.4 ± 8.1 2.07 ± 0.23 Total mean ± SE 1 339 ± 119 80.3 ± 7.1 1.53 ± 0.06 1 627 ± 130 97.6 ± 7.8 1.88 ± 0.1
creatic and gut hormones. In obesity, serum leptin levels are higher than in lean subjects and this divergence is not translated into an early satiety during the meal. Furthermore, leptin has been shown to be involved in vascular smooth muscle cells func-tion and plaque formafunc-tion [23].
In the management of obesity, physical training is an absolute priority together with hypocaloric diet. Nevertheless, debate exists on type of training and intensity of exercise required to accomplish such purpose.
We evaluated in 2 groups of obese subjects the effects of 2 exer-cise training (aerobic alone vs. aerobic with a bout of anaerobic work, 5 min at 85 % of HRmax) performed during a 4-week period. The total amount of work was significantly different when expressed as Kjoules, but similar when related to FFM.
The differences between aerobic and anaerobic work are consid-erable. In fact, aerobic work requires an amount of energy com-patible with the kinetics of the refurnishing process and ATP is restored by means of oxidative mechanisms consuming oxygen and utilizing the body stores of carbohydrates and fats as sub-strate. In such case, physical exercise can be maintained for sev-eral minutes in a steady-state condition. Alternatively, anaerobic work outbalances the capability of the mechanisms for energy restoration so that it may be supported for a limited period by the ATP sources at muscular level. These differences involve dif-ferent dynamics in some chemical endogenous mediators which play a role in metabolism during and after physical exercise, like lactic acid, catecholamines, GH, and others.
Table 5 Individual work performed at beginning and ending training of Group B and simulated data of aerobic work instead of aerobic plus anaerobic work. Group B
Males
Single session work outputs at the beginning of training
Simulated aerobic single session work-outputs at the beginning of training
Single session work outputs at the end of training Watts ( > AT) Kjoules ( > AT) Kjoules/ FFM ( > AT)
Watts Kjoules Kjoules/FFM Watts
( > AT) Kjoules ( > AT) Kjoules/ FFM ( > AT) 1 2 345 (595) 141 (36) 1.65 (0.42) 2 100 126 1.47 2 605 (680) 156.3 (40.8) 1.74 (0.45) 2 * 2 865 (765) 172 (46) 1.75 (0.46) 2 520 151 1.54 3 300 (850) 198.0 (51.0) 2.11 (0.54) 3 2 345 (595) 141 (36) 1.93 (0.49) 2 100 126 1.73 2 780 (680) 166.8 (40.8) 2.23 (0.55) 4 1 735 (510) 104 (31) 1.64 (0.48) 1 470 88 1.39 1 735 (510) 104.1 (30.6) 1.67 (0.49) mean ± SE 2 323 ± 231 139 ± 14 1.74 ± 0.07 2 048 ± 216 123 ± 13 1.53 ± 0.07 2 705 ± 325 156 ± 20 1.94 ± 0.14 (616 ± 53) (37 ± 3) (0.46 ± 0.02) (689 ± 69) (41 ± 4) (0.51 ± 0.02) Females 5 2 085 (510) 125 (31) 2.50 (0.61) 1 890 113 2.27 2 260 (510) 135.6 (30.6) 2.64 (0.60) 6 1 475 (425) 89 (26) 1.70 (0.49) 1 260 76 1.45 1 910 (510) 114.6 (30.6) 2.29 (0.61) 7 1 735 (510) 104 (31) 1.90 (0.56) 1 470 88 1.61 1 910 (510) 114.6 (30.6) 2.16 (0.58) 8 1 650 (425) 99 (26) 1.36 (0.35) 1 470 88 1.21 1 910 (510) 114.6 (30.6) 1.62 (0.43) mean ± SE 1 736 ± 128 104 ± 8 1.87 ± 0.24 1 523 ± 132 91 ± 8 1.64 ± 0.23 1 998 ± 88 120 ± 5 2.18 ± 0.21 (468 ± 25) 28 ± 2 (0.5 ± 0.06) (510) (31) (0.56 ± 0.04) Total mean ± SE 2 029 ± 165 122 ± 10 1.8 ± 0.12 1 785 ± 154 107 ± 9 1.58 ± 0.11 2 301 ± 194 138 ± 12 2.01 ± 0.13 (542 ± 39) (33 ± 2) (0.48 ± 0.03) (595 ± 45) (36 ± 3) (0.53 ± 0.02)
Work beyond anaerobic threshold (AT) are shown in parentheses
Table 6 Measured indexes before and after training in the 2 groups.
Group A Group B
Variable† Training Mean SE p-Value* Mean SE p-Value*
VO2max (ml/min/kg) Before 18.0 (78 %) 1.3 18.8 (76 %) 1.7
After 19.6 (80 %) 1.4 NS 20.1 (78 %) 1.8 NS VO2AT (ml/min/kg) Before 11.9 1.2 12.1 1.5 After 12.8 1.3 NS 14.0 1.6 NS Leptin (μg/l) Before 27.8 6.1 24.2 5.2 After 12.2 2.1 0.04 18.5 3.6 NS LeptinFM (μg/l/kg) Before 0.76 0.1 0.52 0.1 After 0.37 0.1 0.039 0.44 0.1 NS
Insulin (pmol/l) Before 65.2 5.1 72.1 11.8
After 57.8 7.0 NS 73.0 7.1 NS
Glucose (mmol/l) Before 4.89 0.5 4.93 0.2
After 5.07 0.3 NS 4.79 0.3 NS
HOMA2-B Before 112.1 4.6 118.9 11.0
After 90.6 4.9 0.015 134.5 14.3 NS
NEFA (mmol/l) Before 680 84 510 43
After 480 57 0.05 702 81 0.018
† Values are mean ± SEM. NS: Not significant *By two-tailed analysis of variance
Theoretical values in percent are given in parentheses
Our previous comparative studies between different models of physical training demonstrated that the addition of single bouts of exercise beyond the AT to aerobic activity may evoke a sig-nificant measurable GH responsiveness in obesity when exercise intensity exceeds AT [14]. Therefore, the higher levels in GH and catecholamines beyond AT [14, 24] involve an increased lipolytic activity and a more evident FM loss [15]. Aerobic plus anaerobic work does not improve glucose metabolism and increases circu-lating NEFA [15].
Data from the literature suggest that physical exercise reduces serum leptin after a single training at moderate intensity and long duration in normal subjects [9], as well as after prolonged periods of activity at different intensities in sedentary over-weight and obese men [11], and in women after intensive swim-ming training [25]. The decrease of leptin after these long periods of training has been associated to the loss of FM by the authors.
Our study seems to point out the following considerations. In the obese subjects of this study, leptin decreases after a period of training of 4 weeks together with FM, and intensity of work seems to condition the behaviour of both FM and leptin. In fact, aerobic exercise promotes a slight FM loss together with a sig-nificant lowering of leptin, while aerobic exercise associated to a bout of work beyond AT promotes a more important FM loss without a significant reduction of leptin, even when adjusted to the loss in FM.
Some factors are able to contrast the extent of leptin reduction linked to physical stress. One of these could be GH, whose ability to increase serum leptin has been demonstrated in GH deficient subjects [6] and whose increase has been shown in obesity dur-ing a progressive workdur-ing test after traindur-ing with bouts of work beyond AT [14]. The appreciable lowering in FM together with the increased values of NEFA after training in Group B seems to be in agreement with a high, consistent mobilisation of lipids which probably exceeds their dynamic utilisation, linked to the enhanced GH response [15].
Another factor involved herein might be represented by the increase in intracellular products from glucose metabolism caused by the stress from bouts of work beyond AT [8] in absence of improvement in insulin sensitivity for glucose. Hexosamine biosynthesis, a relatively minor branch of glycolysis, has been
observed to regulate leptin production in rodent as well as in human adipose tissue [8]. In fact, culturing human subcutane-ous adipocytes with glucosamine, a relevant increase in leptin release has been detected [26]. Although causal relationships have not been established, numerous studies have shown a cor-relation between increased hexosamine biosynthesis pathway and insulin resistance [27, 28].
Moreover, data from McClain [8] are consistent with a role for muscular tissue in the production of leptin. The higher FFM of our Group B may be in agreement with its leptin behaviour. In our opinion, of some interest may be the linear correlation between the total amount of work performed during the 4-week period and the variations after training of leptinFM (leptinFM
after training/leptinFM before training) as percent, evident in
Group A.
In Group B, the percentages of variation after training of leptinFM
(leptinFM after training/leptinFM before training) demonstrated
an increasing trend and corresponded in 3 cases to the highest total performances. This might be related to the hypothesised increase of hexosamines (which stimulate leptin production) probably linked to anaerobic metabolism and to the increased production of GH as above mentioned. In these cases, the high-est secretions of hexosamines and GH probably overpower the decrease of leptin due to exercise. The heavier FFM of Group B (consistent with a role for muscular tissue in the production of leptin) [9] might be a further factor.
The data of Group A may confirm the relevance of the quantity of work performed, but only aerobic work, in lowering serum lep-tin and may represent a predictive index in a program for con-trolling FM loss, and probably an improvement of satiety with exercise at low intensity.
Some caveats have to be considered as potential limitations of this study, such as the lack of measure of satiety collected before and after workout, the absence of measures of subcutaneous and visceral adiposity, with the former being a greater source of cir-culating leptin than the latter adipose compartment, and the limited number of analysed subjects.
Notwithstanding these limits, we think that the following con-clusions may be allowed.
Physical training, in the absence of dietary restrictions, decreases leptin over-secretion and FM in obese individuals. The decrease of leptin levels is seemingly conditioned by the intensity and type of work, while appearing dissociated from the amount of FM lost during training. After aerobic training, there is a lower FM loss than that obtained after aerobic plus anaerobic exercise. The significant reduction in leptin levels, when related to the variation of FM, correlates with the amount of aerobic work per-formed, that is, greater the amount of aerobic work perper-formed, stronger the decrement in leptin secretion ( ●▶ Fig. 1).
The opposite correlation in Group B appears to be less interest-ing, where the increased amount of total work performed cor-responded to a lower magnitude in the reduction of leptin. Taken together, our current observations seem to confirm that the approach to antiobesity strategies by means of physical training should contemplate the opportunity to initially prescribe a period of work at aerobic and anaerobic intensity to get a sig-nificant reduction of FM, subsequently followed by sessions of aerobic work alone so as to improve FM loss, ameliorate the metabolic profile, reduce lower circulating leptin, and likely to improve satiety control. Nevertheless, a more extended study concerning the behaviour of leptin after aerobic and aerobic plus anaerobic training is desirable to confirm our data.
Fig. 1 Linear regression of the ratio of leptinFM after training and leptinFM before training in percent vs. total amount of performed work, in both groups. * Outlier of group B.
Acknowledgements
▼
The authors wish to thank Dr. Gillian Walker for kindly supervis-ing the text.
Conflict of Interest
▼
The authors declare that they have no conflicts of interest in the authorship or publication of this contribution.
Affiliations
1 Division of Respiratory Rehabilitation, H. San Giuseppe, Istituto Auxologico Italiano, Verbania, Italy
2 Division of Rehabilitation Medicine, H. San Giuseppe, Istituto Auxologico Italiano, Verbania, Italy
3 Department of Translational Medicine, University of Piemonte Orientale ‘A. Avogadro’, Novara, Italy
4 Division of General Medicine, H. San Giuseppe, Istituto Auxologico Italiano, Verbania, Italy
References
1 Considine RV, Caro JF. Leptin and the regulation of body weight. Int J Biochem Cell Biol 1997; 29: 1255–1272
2 Myers MG Jr, Heymsfield SB, Haft C, Kahn BB, Laughlin M, Leibel RL,
Tschop MH, Yanovski JA, and the attendees NIH conference “Toward a Clinical Definition of Leptin Resistance”. Defining clinical Leptin
resist-ance – Challenges and opportunities. Cell Metab 2012; 15: 150–156 3 Teta D, Bevington A, Brown J, Throssel D, Harris KPG, Walls J. Effects
of acidosis on leptin secretion from 3T3-L1 adipocytes and on serum leptin in the uraemic rat. Cli Sci 1999; 97: 363–368
4 Mueller WM, Stanhope KL, Gregoire F, Evans JL, Havel PJ. Effects of metformin and vanadium on leptin secretion from cultured rat adi-pocytes. Obes Res 2000; 8: 530–539
5 Mueller WM, Gregoire FM, Stanhope KL, Mobbs CV, Mizuno TM,
War-den CH, Stern JS, Havel PJ. EviWar-dence that glucose metabolism regulates
leptin secretion from cultured rat adipocytes. Endocrinology 1998; 139: 551–558
6 Gill MS, Toogood AA, Jones J, Clayton PE, Shalet SM. Serum leptin response to the acute and chronic administration of growth hormone (GH) to elderly subjects with GH deficiency. J Clin Endocrinol Metab 1999; 84: 1288–1295
7 Boden G, Chen X, Mozzoli M, Ryan I. Effect of fasting on serum leptin in normal human subjects. J Clin Endocrinol Metab 1996; 81: 3419–3423 8 McClain DA, Alexander T, Cooksey RC, Considine RV. Hexasamines
stimulate leptin production in transgenic mice. Endocrinology 2000; 141: 1999–2002
9 Wang J, Liu R, Hawkins M, Barzilai N, Rossetti L. A nutrient-sensing pathway regulates leptin gene expression in muscle and fat. Nature 1998; 393: 684–688
10 Nindl BC, Kraemer WJ, Arciero PJ, Samatallee N, Leopne CD, Mayo MF,
Hafeman DL. Leptin concentrations experience a delayed reduction
after resistance exercise in men. Med Sci Sports Exercise 2002; 34: 608–613
11 Zaccaria M, Ermolao A, Brugin E, Bergamin N. Plasma leptin and energy expenditure during prolonged, moderate intensity, treadmill exercise. J Endocrinol Invest 2013; 36: 396–401
12 Plonka M, Toton-Morys A, Adamski P, Suder A, Bielanski W, Dobrzanska
MJ, Kaminska A, Piorecka B, Glodzik J. Association of the physical
activ-ity with leptin blood serum level, body mass indices and obesactiv-ity in schoolgirls. J Physiol Pharmacol 2011; 62: 647–656
13 Guelfi KJ, Donges CE, Duffield R. Beneficial effects of 12 weeks of aerobic compared with resistance exercise training on perceived appetite in previously sedentary overweight and obese men. Metabolism 2013; 62: 235–243
14 Salvadori A, Fanari P, Marzullo P, Codecasa F, Tovaglieri I, Cornacchia
M, Wolker G, Brunani A, Longhini E. Dynamics of GH secretion during
incremental exercise in obesity, before and after a short period of training at different work-loads. Clin Endocrinol 2010; 73: 491–496 15 Salvadori A, Fanari P, Marzullo P, Codecasa F, Tovaglieri I, Cornacchia
M, Brunani A, Luzi L, Longhini E. Short bouts of anaerobic exercise
increase non-esterified fatty acids release in obesity. Eur J Nutr 2014; 53: 242–249
16 Lukaski HC, Bolonchuk WW, Hall CB, Siders WA. Validation of tetrapolar bioelectrical impedance method to assess human body composition. J Appl Physiol 1986; 60: 1327–1332
17 Harris DJ, Atkinson G. Update-ethical standards in sport and exercise science research. Int J Sport Med 2011; 32: 819–821
18 Wasserman K, Hansen JE, Sue DY, Casaburi R, Whipp B. Principles of exercise testing and interpretation. 3rd edn. Philadelphia: Lippincott, Williams & Wilkins, 1999; 65–67
19 Beaver WL, Wasserman K, Whipp BJ. A new method for detecting anaer-obic threshold by gas exchange. J Appl Physiol 1986; 60: 2020–2027 20 Wallace TM, Levy JC, Mattews DR. Use and abuse of HOMA modelling.
Diabetes Care 2004; 27: 1487–1495
21 Steel RGD, Torrie JH. Principles and procedures of statistics, a biometri-cal approach. McGraw-Hill, New York: 1980; 188–190
22 Kleinbaum DG, Kupper LL, Muller KE. Applied regression analysis and other multivariate methods. Boston: PSW-Kent Publ, 1988; 49–53 96–99, 262–276
23 Tesauro M, Cardillo C. Obesity, blood vessels and metabolic syndrome. Acta Physiol 2011; 203: 279–286
24 Salvadori A, Fanari P, Giacomotti E, Palmulli P, Bolla G, Tovaglieri I, Luzi
L, Longhini E. Kinetics of catecholamines and potassium, and heart
rate during exercise testing in obese subjects. Eur J Nutr 2003; 42: 181–187
25 Hayase H, Nomura S, Abe T, Izawa T. Relation between fat distribu-tions and several plasma adipocytokines after exercise training in premenopausal and postmenopausal women. J Physiol Anthropol 2002; 21: 105–113
26 Considine RV, Cooksey RC, Williams LB, Fawcett RL, Zhang P, Ambrosius
WT, Whitfield RM, Jones RM, Inman M, Huse J, Whitfield RM, McClain DA. Hexosamines regulate leptin production in human subcutaneous
adipocytes. J Clin Endocrinol Metab 2000; 85: 3551–3556
27 Emilsson V, O’Dowd J, Nolan AL, Cawthorne MA. Hexosamines and nutrient excess induce leptin production and leptin receptor activa-tion in pancreatic islets and clonal B-cells. Endocrinology 2001; 142: 4414–4419
28 Buse MG. Hexosamines, insulin resistance, and the complications of diabetes: current status. Am J Physiol Endocrinol Metab 2006; 290: E1–E8
